RAM: DDR vs DDR2

[Katicius]

Hello, I've got 2GB of Kingston Ram in dual channel mode (DDR400). As I heard, Kingston memory is the best, it can be overclocked perfectly, so I dedided to OC my Kingston. I've increased the frequency by 60MHz (460MHz), tested it with PCMark05 memory test and got 4684 points:
http://img137.imageshack.us/img137/8421/ramyv5.jpg

According to Tom's Hardware Guide, DDR2 - 800MHz Crucial RAM scores in the same test 4487 with C2D EX6800 CPU.
http://www23.tomshardware.com/cpu.html?modelx=33&model1=430&model2=464&chart=172

My PC Spec:
Intel Pentium 4 Extreme Edition 3,5 GHz (Dual SMT, 1GHz FSB),
GeForce 7800GS+ (G71) (7900GT)
2 GB of Kingston RAM
Seagate Barracuda 720GB SATA2, NCQ, 7200/16MB.

What do you think about my test, was it right?
Waiting for your answers, thanks

 

[neocristi]

Hello, I've got 2GB of Kingston Ram in dual channel mode (DDR400). As I heard, Kingston memory is the best, it can be overclocked perfectly, so I dedided to OC my Kingston. I've increased the frequency by 60MHz (460MHz), tested it with PCMark05 memory test and got 4684 points:
http://img137.imageshack.us/img137/8421/ramyv5.jpg

According to Tom's Hardware Guide, DDR2 - 800MHz Crucial RAM scores in the same test 4487 with C2D EX6800 CPU.
http://www23.tomshardware.com/cpu.html?modelx=33&model1=430&model2=464&chart=172

My PC Spec:
Intel Pentium 4 Extreme Edition 3,5 GHz (Dual SMT, 1GHz FSB),
GeForce 7800GS+ (G71) (7900GT)
2 GB of Kingston RAM
Seagate Barracuda 720GB SATA2, NCQ, 7200/16MB.

What do you think about my test, was it right?
Waiting for your answers, thanks

In order to measure RAM performance, you have to do a read/write test. Use Sisoft Sandra and measure RAM bandwidth.

By the way... Kingston HyperX are very good modules. Your 3 3 8 timings tell me that you don't have the HyperX right?


Anyway, don't worry so much about RAM. Quantity is way more important than speed or timings.

 

[Mondoman]

It's always fun to see how much you can boost various test results by OCing! That said, benchmark tests never quite seem to capture all the complexity of real-world apps, so you might want to try running some app-based tests (for example, frame rates on your favorite game, etc).